Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 21-GEMMASCOPE-RES-16K
    6. 6763
    Prev
    Next
    INDEX
    Explanations

    technical specifications related to video encoding and formats

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    djangoproject
    -0.60
    initializeApp
    -0.55
    AxisAlignment
    -0.55
     typelib
    -0.54
    LLocation
    -0.53
    úncia
    -0.52
    actionMode
    -0.51
    EDEFAULT
    -0.51
     Diabetes
    -0.50
     unread
    -0.50
    POSITIVE LOGITS
     video
    1.10
     Video
    1.08
    Video
    1.07
    video
    1.02
     vidéo
    0.95
     VIDEO
    0.85
    视频
    0.82
     videos
    0.82
    VIDEO
    0.80
     vídeo
    0.78
    Activations Density 0.342%

    No Known Activations